Geographic data is information that describes both the location of something on Earth and its characteristics. It combines two parts: spatial data (where something is) and attribute data (what that something is). Every pin on a digital map, every satellite image of a forest, and every GPS coordinate logged by your phone is a form of geographic data. It underpins everything from turn-by-turn navigation to urban planning, and the global market for systems that manage it was valued at $9.4 billion in 2024.
The Two Parts: Location and Attributes
Every piece of geographic data has a spatial component and an attribute component working together. The spatial component tells you where something exists, usually as coordinates (latitude and longitude). The attribute component tells you what’s there and describes its properties. A point on a map marking a hospital, for example, carries spatial data (its coordinates) and attribute data (its name, number of beds, hours of operation).
Neither part is useful alone. Coordinates without context are just numbers. A list of hospital names without locations is just a directory. Geographic data becomes powerful when the two are linked, letting you ask questions like “which hospitals within 10 miles have an emergency department open right now?”
Vector vs. Raster: Two Ways to Model the World
Geographic data is stored in one of two fundamental formats, and each represents the world differently.
Vector data uses points, lines, and polygons to represent features. A city might be a point, a highway a line, a national park a polygon. Vector data is precise. It can represent the exact boundary of a property or the exact path of a river, which is why it’s the standard for maps where accuracy matters, like parcel boundaries or road networks.
Raster data divides the world into a grid of cells (pixels), where each cell holds a value. Satellite imagery is raster data. So is a land-cover map that colors each cell as forest, water, or urban area. Raster works well for continuous phenomena like elevation, temperature, or vegetation density, where every location has a value rather than a sharp boundary. One of the most common raster datasets is land-cover classification derived from Landsat satellite imagery.
There’s an old saying in the field: “raster is faster, but vector is corrector.” Raster data is computationally simpler to process, but vector data captures shapes and boundaries with greater precision. Most real-world projects use both.
How Geographic Data Gets Collected
The methods for gathering geographic data range from handheld GPS units to satellites orbiting hundreds of miles overhead.
GPS receivers are the most familiar tool. A standard civilian GPS unit today achieves a median horizontal accuracy of about 3 meters and vertical accuracy of about 3.6 meters. That’s precise enough for navigation and basic mapping, though professional survey equipment can push accuracy down to centimeters using correction signals.
Satellite and aerial imagery captures broad areas at once. Multispectral sensors on satellites like Landsat record light across wavelengths the human eye can’t see, which lets analysts distinguish healthy vegetation from stressed crops or map water bodies from space.
LiDAR (Light Detection and Ranging) is a laser-based technology typically mounted on aircraft. It fires rapid pulses of light at the ground and measures how long each pulse takes to bounce back, building a dense “point cloud” of elevation measurements. LiDAR achieves vertical accuracy within about 10 centimeters (4 inches), making it the go-to tool for flood modeling, forestry inventories, and terrain mapping. In places like Alaska, where persistent cloud cover blocks optical sensors, a radar-based alternative called Interferometric Synthetic Aperture Radar fills the gap.
Ground surveys still matter for high-stakes measurements like property boundaries and construction sites, where centimeter-level precision is non-negotiable.
Coordinate Systems and Datums
Raw coordinates are meaningless without a reference framework. A coordinate system defines how positions on Earth’s curved surface translate into numbers, and a geodetic datum provides the mathematical model of Earth’s shape that those numbers are based on.
Think of it this way: Earth isn’t a perfect sphere. It’s slightly flattened at the poles and lumpy with mountains and ocean trenches. A datum approximates this shape with an ellipsoid, a smoothed-out mathematical surface, and defines where “zero” is for latitude, longitude, and elevation. The most widely used datum today is WGS 84, the reference system behind GPS. For vertical measurements in North America, the standard is NAVD 88, which defines elevations relative to sea level.
Using mismatched datums is one of the most common sources of error in geographic data. Two datasets can describe the exact same location but show it in slightly different places if they’re built on different reference systems. Converting between datums is routine, but forgetting to do it can shift features by meters or even hundreds of meters.
Common File Formats
Geographic data gets stored in a variety of formats, each with trade-offs.
- Shapefile: One of the oldest and most widely exchanged vector formats. Despite its name, it’s actually a bundle of several files (.shp, .shx, .dbf at minimum) that must travel together. Nearly every GIS tool can read shapefiles.
- GeoJSON: A lightweight format for vector features based on JSON, the same structure used across web development. It’s the natural choice for web maps and browser-based applications.
- KML: Originally developed by the company behind Google Earth, KML became an open standard in 2008. It’s XML-based and commonly used for sharing map overlays. When bundled with images or other files, it’s compressed into a KMZ.
- GeoPackage: A newer format designed by the Open Geospatial Consortium to work across devices, from laptops to phones. It stores vector features, tables, and raster tiles in a single portable file.
- Raster formats: Standard image formats like TIFF, JPEG, and PNG all serve as raster containers. GeoTIFF is especially common because it embeds coordinate information directly in the file.
For larger operations, spatial databases like PostGIS (an extension of PostgreSQL) and SpatiaLite store geographic data alongside traditional database records, enabling complex queries across millions of features.
Static Data vs. Real-Time Streams
Traditional geographic data is static: a land-cover map, a census boundary file, a topographic survey. You collect it, process it, and it represents a snapshot in time. But a growing share of geographic data is dynamic, streaming in continuously from sensors, vehicles, and connected devices.
Real-time geographic data comes from sources like GPS trackers on delivery trucks, air-quality sensors mounted on buildings, or IoT devices monitoring water levels in storm drains. These feeds emit location-tagged observations at high frequency, sometimes multiple times per second. Systems that ingest this data can trigger automated alerts when patterns emerge, like rerouting emergency vehicles when a flood sensor trips, or notifying a logistics dispatcher when a shipment deviates from its planned route.
The distinction matters because real-time data requires different infrastructure. Static data can sit in a file and be analyzed whenever you need it. Streaming data needs to be ingested, processed, and acted on as it arrives, then optionally stored for later analysis.
Where Geographic Data Gets Used
Geographic data touches more industries than most people realize. Logistics companies use it to optimize delivery routes, reduce fuel costs, and track assets in real time. Urban planners rely on it to analyze traffic patterns, manage waste collection, assess environmental impacts, and decide where to build new infrastructure. Emergency management teams use it to model community risks, visualize disaster scenarios, and find the fastest routes to incidents.
In public health, geographic data maps disease outbreaks and identifies underserved populations. In agriculture, it guides precision farming by showing which parts of a field need more water or fertilizer. Insurance companies use it to price flood and wildfire risk. Retailers use it to choose store locations based on population density and competitor proximity. The common thread is that nearly every decision improves when you can see where things are and how they relate spatially.
Privacy and Location Data
Location data tied to individuals carries serious privacy implications. Your phone’s GPS log, your check-in history, or your car’s telematics data can reveal where you live, work, worship, and seek medical care. Under regulations like the EU’s General Data Protection Regulation, location data tied to an identifiable person counts as personal data. That means organizations collecting it must be transparent about how they use it, give individuals the right to access and delete their data, and obtain clear consent before processing it.
Anonymizing location data is harder than it sounds. Even stripped of names, a dataset of movement patterns can often be re-identified because most people follow unique daily routines. Effective anonymization typically requires aggregating data so that individual tracks dissolve into crowd-level patterns, sacrificing precision to protect privacy.